A distributed capacity estimation algorithm of cognitive network

The capacity analysis of cognitive network is rarely studied as the dramatic dynamic character of opportunistic spectrum usage. As the optimal result of network capacity is NP complete, it is necessary to design an efficient algorithm to find the approximated optimal results. This paper proposed a distributed algorithm which is based on the minimum clique partition of the network´s contention graph. Based on the clique partition result of the network, the capacity of the network is easy to find out. We have proved that the computer complexity of this algorithm is O(n³). By detailed simulation, the performance of this algorithm is closely to the optimal results.